220
Jan. 20, 1963
COMMUNICATIONS T O THE EDITOR COORDINATED CATIONIC INITIATOR FOR HIGH cZS-I,~-POLYBUTADIENE
Sir:
Evidence presented in this communication indicates that in the case of soluble cobalt initiators1v2for polymerizing butadiene to a high cis-1,4 configuration the propagation step occurs through an intermediate carbonium ion rather than a cobalt-carbon bond.a Carbon-14 methanol and butanol tritiated on the oxygen were used to stop active butadiene polymerizations initiated by (C2H6)BA12Cla-cobalt octoate. As shown by Feldman and Perry4 carbon-14 would be bonded to the polymer if a carbonium ion were involved in the chain growth process while tritium would be bonded to the polymer if a carbanion were involved. The C14 methanol was purchased from New England Nuclear Corporation and two ml. of tritiated butanol was prepared by an exchange reaction with 0.01 ml. of tritiated water. About 0.2 g. of Linde Molecular Sieve 5A then was added to remove most of the water. In the polymerization experiments 0.005 millimole of cobalt octoate, 0.05 millimole of (C2H6)3A12C13 and 150 milliliters of benzene were placed in each of two 250-1111. round-bottom flasks. Butadiene gas then was bubbled into each flask a t the same rate. A rapid temperature rise resulted from the formation of high cis-1,4-polybutadiene. After 60 seconds a t which time the temperature had increased 15’, one polymerization was stopped with 0.1 ml. of tritiated butanol and the second with 0.1 ml. of carbon-14 methanol. Catalyst residues were removed by twice precipitating the soft tacky polymer with methanol and redissolving in benzene. Polymer activities which were measured by liquid scintillation counting of toluene solutions containing PPO (2,5-diphenyloxazole) and POPOP [2,2-pphenylenebis(5-phenyloxazole) ] were Shortstop
CaHsOH8 ‘‘CH30H
Shortstop activity (millicuries)
Polymer activity (microcuries/g.)
0.04 0.05
0.008 0.96
The small tritium activity was about equal to the background count and can be explained easily by residual moisture in the butanol. Carbon-14 activity, however, can only arise from active carbonium ions on the polymer chains. Assuming that all polymer chains were still growing and were terminated by methanol, one calculates a number average molecular weight of 21,000, which is a reasonable figure based on appearance of the polymer. These results then show that the growing polymer ion is a cation and not an anion. Failure3 to find carbon-14 in cis-1,4-polybutadiene prepared with (14C~H.&A1C1-cobaltbisacetylacetonate is thus explained since metal-carbon bonds are most probably not involved. The growing carbonium ion may be associated with a complex anion containing the metals. UNITEDSTATESRUBBERCOMPANY RESEARCH CENTER C. W, CHILDERS WAYNE,NEWJERSEY RECEIVED NOVEMBER 15. 19G2 (1) C. Longiave, R. Castelli and G . F. Croce, L a Chimica E L’lnduslrim 18, 625 (1961). (2) M . Gippin, I & E C Prod. Res. and Deuelcg., 1. 32 (1962). (3) G . Natta, L. Porri and A. Carbonaro, Rendiconti, 29, 491 (1960). (4) C. F. Feldman and E. Perry, J . Polymer Sci., 46, 217 (1960).
PHASE DEPENDENCE OF CARBON-11 RECOIL PRODUCTS IN ETHANE AND PROPANE ; EVIDENCE FOR METHYLENE INSERTION’
Sir: The recoil product spectrum from ethane in the gas and liquid phases and from propane in the gas, liquid and solid phases shows definite phase dependence.2 The data provide added evidence for the importance of energy redistribution and deexcitation of the “complex”8 formed. Further, the products suggested to have been formed by methylene-Cl1 insertion reactions show a predictable yield trend on changing phase, lending additional support for the postulated existence of this reactive The C12(n,2n)C11 reaction was used to produce the carbon-11 atoms. The C12(p,pn)C11 reaction was used for absolute yield determinations in the gas and liquid phases. The experimental irradiation technique for the C12(n,2n)C11 reaction was essentially that described by Suryanarayana and Wolf .s The radiation dose delivered to the gaseous samples was 1.4 X ev./molecule. The absorbed dose in the liquid and solid was higher for the C12(n,2n)C11reaction and a factor of 10 lower for all comparable runs involving the C12(p,pn)C11 reaction. Phillips research grade alkanes further purified by multiple crystallization were used in this work. Analysis of the carbon-11 labeled compounds was done by radio-gas chromatography The change from gas to liquid phase causes a drop in the sum of ethylene and acetylene yields (Table I) in both ethane (55 t o 33%) and propane (43 to 31%). Lang and Voigt7 also observed a marked decrease in ethylene and acetylene from cyclohexane [C1z(y,n)C1l] as the temperature of the sample being irradiated was decreased from - 30 to - 78’. A further decrease in these yields is observed for solid propane. Insofar as all or part of the acetylene is formed by carbon atom insert i ~ n , followed ~ , ~ ~ by fragmentation of the energy-rich “ ~ o m p l e x ”(whether ~ this complex be a discrete intermediate such as an insertion product or whether it be in some more loosely bonded initial state) attenuation of this energy by the surroundings should tend to decrease this degradation provided, (1) the energy of the first formed complex is not too high and, (2) the rate of deexcitation and the rate of internal rearrangement of the “complex” is of the same order of magnitude as the fragmentation reaction. A concomitant increase in three carbon compoundsforethane and four carbon compounds ( 1 ) Research performed under the auspices of the U. S. Atomic Energy Comrnisssion. ( 2 ) C f . C. MacKay and R. Wolfgang, Rcdiochimica A c t a , 1, 42 (1962). A preprint of this article “Phase Independence of Major Reaction Mechanisms of Recoil Carbon Atoms” was sent to us by the authors. They studied ethylene, ethylene oxide and isobutane (3) C f . J. Yang and A. P . Wolf, J . A m . Chem. SOC.,8 2 , 4488 (1960). (4) (a) A. P . Wolf, B. Gordon and R . C. Anderson, ibid., 78, 2657 (1956). (b) A. P . Wolf. C. S. Redvanly and R . C. Anderson, :bid., 79, 3717 (1957). (c) C. MacKay and W. F. Libby, ibid., 79, 6366 (1957). (d) B. Gordon and A. P . Wolf, Brookhaven National Laboratory Annual Report, BNL-523, AS-12.51, 1958. (e) Cf.reference 5. (f) A. P. Wolf, A n n . Rev. N u d . Chem., 10, 259 (1960). (9) A. P . Wolf, “Chemical Effects of Nuclear Transformations,” IAEA, Vienna, 1961, Vol. 11. (h) C. MacKay, M . Pandow, P . Polak and R . Wolfgang, ibid., reference 4g. (5) B. Suryanarayana and A. P. Wolf, J. P h y s . Chem., 81, 1369 (1958). ( 6 ) Cf.F. Cacace, Nucleonics. 19, no. 5, 45 (1961), for a review of this technique. Specific techniques used in this work will be found in G. Stocklin, F. Cacace, and A. P . Wolf, Z . A n a l . Chem., in press. (7) C. E. Lang and A. F. Voigt, J . P h y s . Chem., 66, 1542 (1961).
COMMUNICATIONS TO THE EDITOR
230
CARBOX-11 CONTAINIXG PRODUC’TS KESULTING FKOM
TABLE I C1’(n,2n)C”
THE
hACTIOK
Et hanea--
Compound
Carbon monoxide
Gas 1 atm. 25’
Liquid
- 166’
